The Seattle Obesity Study (S.O.S) seeks to explain why obesity rates are more prevalent among some racial/ethnic minorities, population subgroups of lower education and incomes, and in high-poverty areas. The study will recruit a longitudinal cohort of 400 male and female adults, ages 21-65y, and resident in Seattle-King County and stratified by income and race/ethnicity, using an address-based sampling scheme. Data on food seeking behavior will be collected using a combination of self report and objective measures. At the first in- person interview, participants will complete a behavior survey module on food shopping and dietary habits, estimated food expenditures, nutrition literacy, attitudes towards diet and health, demographics, and health status. Health-related questions will address obesity, diabetes, stress, and cardiovascular disease. Height, weight, and waist circumference will be measured. A GPS receiver with data logging capability will be used to track the participants' movements and to record the time and location of all food-related events for 1 wk. Food store and restaurant receipts, stamped with time and date, will be collected for 4 wks. Foods brought into the home will be scanned with a portable UPC scanner for 4 wks. Three 24-h food recalls will be administered by phone in wks 3 and 4, and food frequency questionnaires will be administered in week 4 of data collection. To be able to establish causality between environmental and dietary variables and health outcomes, the behavior survey module and the food frequency questionnaire will be re-administered during a follow up visit 18 months later and anthropometric measures will be collected for the second time. Data on foods purchased for use at home and foods consumed away from home will be linked to individual-level metrics of the built environment based on tax parcel data. The main aim of the study is to assess the impact of SEP, environmental, and dietary variables on obesity and body weight gain. The hypothesis is that low dietary quality and low diet cost will be associated with prevalent obesity and will predict 18 mo. weight gain, adjusting for SEP. This model will be built on data collected and validated under Aims 2 and 3. The second aim will be to assess how access to food sources (both physical and economic), retail outlet choice, and food purchases impact diet quality across SEP strata. The hypothesis is that diet quality will be influenced strongly by price and other economic variables than by proximity to food sources. The third aim will be to validate self-reported data on food acquisition and expenditures against objective measures of individual behavior in space and time. The hypothesis is that objective measures of food purchase behaviors based on GPS logs, food store and restaurant receipts, and UPC scanner data will provide new insights into the energy and nutrient content of foods that are acquired at different locations. Study results have the potential of informing local and national policies on improving retail access to affordable healthy foods by vulnerable groups.